A Case Study in Fostering a Learning Culture in the Context of COVID-19

The surge of COVID-19 patients during the global pandemic required hospitals and providers to create new clinical, operational, and staffing protocols that extended capabilities beyond conventional standards of care. The evolving science of the disease, coupled with worldwide disruptions to supply chains and limited availability of scarce resources, amplified the challenges of sustaining organizational discipline in a highly dynamic environment.

Maintaining high-reliability systems — especially during a pandemic — requires hospital leaders and frontline managers to continually be mindful of the systems and processes affecting patient care.(1) Being preoccupied with failure, avoiding the temptation to over-simplify problems, being sensitive to dynamic conditions, fostering staff resilience, and deferring to frontline expertise are the hallmarks of high-reliability organizations.(2) Although the COVID-19 pandemic exposed operational and logistical fractures, it also inspired high-reliability organizations to develop new operational tools and processes for real-time crisis management.

Identifying and incorporating lessons learned as crises unfold is challenging. Organizations often wait until the end of a crisis to conduct “after-action reviews” (AARs) and reflection.(3) While conventional after-action review tools such as the Army AAR process are useful, application of these tools tends to be infrequent and is often triggered by event-based errors rather than continuous learning throughout the lifecycle of a project.(4) Identifying lessons learned during a crisis promotes real-time feedback by capturing intermediate insight.

To address interim learning and catalog important project reflections, NASA developed a Pause and Learn (PaL) action-learning process.(4) NASA uses the PaL process throughout the duration of its long missions. Because PaL features a simple design with targeted and timely feedback, Holy Cross Health adapted NASA’s PaL process to capture lessons learned during its first 16-weeks response to COVID-19.

COVID-19 Surge Planning and Response

Holy Cross Health is a member of Trinity Health, a 92-hospital system in 22 states, and is a Catholic, not-for-profit health system that includes Holy Cross Hospital, Holy Cross Germantown Hospital, Holy Cross Health Network, and the Holy Cross Health Foundation. Each year, Holy Cross Health serves more than 240,000 individuals from Maryland’s two largest counties, Montgomery and Prince Georges, which represented more than half of all COVID-19 cases in the state during the first 16 weeks of the pandemic response.(5)

In late February, Holy Cross Health activated its incident management center (IMC) to manage and coordinate the operations, planning, finance, logistics, and communications components of Holy Cross Health’s response to COVID-19. Within the first 16 weeks, Holy Cross Health surged multidisciplinary capabilities, including opening alternate care sites for patient triage; reactivating decommissioned spaces; doubling ICU capacity; creatively sourcing personal protective equipment (PPE); developing new infection control practices; procuring new testing platforms; redeploying and extending staff; and establishing telehealth capabilities at its clinics and for family/visitor connectivity de novo.

While the rapid pace of Holy Cross Health’s transformation led to innovative solutions to address COVID-19, new challenges in sustaining coordinated responses arose because many factors changed simultaneously. Nothing about managing the pandemic has been static. Traditional supply chains failed, guidance on PPE evolved, some equipment that arrived from emergency stockpiles was unusable, and staff extended to their limits as caseloads soared. These problems amplified operational challenges of adapting quickly and implementing new processes in a highly dynamic environment.

At week 16 of Holy Cross Health’s response, the executive leadership team identified the need to take inventory of our actions. Inpatient COVID-19 volumes began to decline, providing a natural point for pause and reflection. Maryland continued to experience an increase in new cases, however, suggesting the possibility of future surges. The president and CEO of Holy Cross Health asked for an internal review of the health system’s response to capture lessons learned. To guide this discussion, the leadership team turned to the Pause and Learn process.

Application of the Pause and Learn Process

The Pause and Learn (PaL) process was developed at the NASA Goddard Space Flight Center to formalize interim lessons learned during the lifecycle of a project.(4) PaL engages in foundational principles of high-reliability organizations (HROs), including deference to expertise and sensitivity to operations.(6,7) PaL bridges the gap between individual and team learning and includes the following features(6):

• Replaces top-down lectures and critiques.

• Gives voice to team members offering views and ideas.

• Is held close in time to the events that happened.

• Involves facilitated, non-attribution participant discussion in safe space.

• Solicits personal point-of-view reflection of events.

• Is conducted periodically following critical events and key milestones throughout a project.

• Is a simple process with minimum intrusion on team workload.

PaL is a tool that can help organizations transform into a “double loop” learning culture, a concept first coined in the late 1970s by Chris Argyris.(8) Whereas most organizations operate in a “single loop” learning environment focusing on addressing actions within a system, double loop learning looks at ways to modify an organization’s underlying norms, policies, and objectives (see Figure 1).(9)

Figure 1. Theories of action — single loop versus double loop learning framework(7)

Double loop learning is particularly useful when leaders need to make informed decisions in rapidly changing and often uncertain contexts.(9) It helps organizations evaluate the reasons they are operating a certain way and how they might change their responses to make the overall system more productive.(9) By reflecting and looking together at the systemic issues, not just addressing individual observations, the PaL was an ideal tool to reframe how Holy Cross Health surged its capabilities and obtained interim feedback from diverse points of view. This aspect of the PaL required the group to meet physically to the extent practicable and in a socially distanced way.

Implementation Methodology

The NASA PaL process was adapted as necessary to fit the healthcare environment but included all the key steps beginning with gathering background observations.(6) Two weeks before the PaL session, 53 leaders and managers solicited feedback from their staff on three questions: Reflecting on the last 16 weeks of our COVID-19 response, (1) What worked well? (2) What didn’t work well? (3) What would you do differently? Responses were not limited to specific domain areas of expertise; individuals were encouraged to provide feedback on any topic they felt relevant for discussion.

The respondents were involved in all aspects of Holy Cross Health’s COVID-19 response, including emergency services; nursing; medical and surgical services; postoperative and intensive care; infection control; behavioral health; facilities management; environmental services; supply chain and logistics; pharmacy; ancillary and support services; discharge planning; volunteer services; visitor and family engagement; chaplain services; colleague health; information technology; human resources; finance; legal and regulatory; ethics; strategy and communications; advocacy; liaisons with local skilled nursing facilities; and our network of community-based clinics at Holy Cross Health.

Responses were submitted to the facilitator in advance of the session. In total, 53 leaders and managers provided more than 150 pages of responses on 56 unique topics. The facilitator tabulated and recorded all responses, noting duplications. Responses were then sorted by topic and classified into five main areas that aligned with the functional organizational structure of the Incident Management Center: (1) clinical operations, (2) planning, (3) logistics, (4) communications, and (5) finance (see Table 1). Each participant in the PaL session received a summarized report for review in advance of the PaL session.

On June 12, 2020, the executive team convened all 53 leaders throughout Holy Cross Health and participated in a two-hour facilitated PaL discussion. This meeting was held in a socially distanced manner with a combination of remote web-based participation and in-person participation in an auditorium.

The facilitator first established ground rules for discussion based on the mission and core values of Holy Cross Health. These rules helped foster collaborative discussions while challenging the team’s mindset and assumptions as they reflected on the first 16 weeks of the surge:

• Be discreet. Nurture a healing, safe environment for all. Unless explicitly stated, what is said in the room stays in the room.

• Be honest. When an activity involves you directly, call it as you see it and proactively participate. Hold ourselves accountable for our actions.

• Be tolerant. Speak with integrity. Honor the dignity of others. Opinions and perspectives are equally important, regardless of speaker’s rank or experience.

• Be a team. When looking at individual actions, view them from the perspective of team responsibility for ensuring excellence and promoting the common good.

Next, the facilitator provided important background to help “level set” the discussion. The facilitator presented a contextual timeline of milestones over the four months that overlaid major highlights of external COVID-19 responses (see Figure 2). The presentation also included Holy Cross Health-specific data on COVID-19 measures, including admissions, discharges/discharge dispositions, and deaths.

Figure 2. Timeline of key events, 2020

The facilitator then reviewed the summary of responses, providing an overview of key findings and noting common themes. Several examples fell into discrete observations of what worked well and where there were notable challenges.

Notable successes included:

• Surged operational capabilities in new ways. Holy Cross Health rapidly adopted telehealth competencies at its clinics where previously there were none. Holy Cross Health devised new methods and approaches to creatively sourcing PPE, including using local procurement channels and engaging community networks to sew gowns from repurposed operating room drapes. Holy Cross Health was an early advocate for the testing and procurement of multiple testing platforms and was a study site for Veklury (remdesivir), for immunotherapy with convalescent serum, and for several other medications designed to block the immune cascade responsible for organ failure and death.

• Developed a mastery of adapt-and-react protocols. Holy Cross Health quickly established patient triage protocols in alternate care sites and adapted these processes as patient volumes surged. Holy Cross Health moved “upstream” in its triage of skilled nursing facilities by coordinating with the state and county health departments to deploy “911 Strike Teams” to local nursing homes to train providers on PPE and to test patients and staff.

• Designed multidisciplinary decision-making processes. To garner support from physicians and staff, Holy Cross Health collaborated with internal and external stakeholders in developing policies and procedures for resuming elective care (i.e., testing protocols, prioritization, block time allocation).

Like many hospitals across the country, Holy Cross Health experienced logistical challenges that affected almost every aspect of patient care. PPE was in short supply. The workforce increased and enhanced its operational tempo to surge critical care capabilities. In late March, when Maryland’s Department of Health required providers to suspend elective care at the height of the state’s initial surge in cases, Holy Cross Health, in accordance with its emergency operations plans, redeployed staff to help address critical needs. Additionally, streamlining communication with patients, providers, and staff presented new challenges, particularly as treatment guidelines and protocols continuously evolved.

Based on the initial feedback, the facilitator focused the group discussion on the four topics that had high response rates and multiple observations from participants, particularly in areas where there were conflicting perspectives. These topics were: (1) PPE/infection control, (2) staffing, (3) logistics, and (4) communications. Facilitation focused on root-cause analysis and included a healthy discussion on proposed recommendations and modifications.

Results and Operational Implications

Feedback was cataloged and synthesized into discrete actions for follow-up. The team developed an inventory of the best practice tools, plans, and protocols that were included in an interim AAR in preparation for future surges.

Participant evaluation of the PaL process was positive. Of the 53 total participants, 31 responded (57 percent response rate) to an eight-question survey based on a 5-point Likert scale (5=strongly agree; 1=strongly disagree) with a free-form comment section.

All respondents either strongly agreed or agreed that the format of the facilitated session worked well (average score 4.6 out of 5.0). Additionally, all respondents either strongly agreed or agreed that the session helped them learn something new (average score of 4.5 out of 5.0). One responder commented, “Pause and Learn should be applied to any lengthy incident — it should be made routine.”

Adapting and implementing the NASA PaL process in a hospital setting was fairly easy. As a best practice, NASA makes use of departmentally neutral facilitators for their PaLs on large space missions. Holy Cross Health adopted a similar approach and used a “neutral” facilitator who was external to the organization for its session, which fostered productive discussions. The facilitator had a background in public health and was integrated into the incident management team at Holy Cross Health.

As a best practice, the facilitator at Holy Cross Health was directly involved in the analysis and synthesis of participant responses in advance of the PaL session. The facilitator was adept at qualitative analysis and organized and documented common themes in a systematic and succinct way. During the discussion, the facilitator was able to keep pace to cover sufficient breadth while allowing space for group reflection and discovery.

Conclusion

The PaL process successfully captured the intermediate lessons learned from Holy Cross Health’s response to COVID-19. Current emergency preparedness frameworks do not include an intermediate evaluation structure for acute or prolonged incidents. PaL helped leaders and frontline staff better understand the reasons and root causes behind what worked well, what did not, and what should be done differently. Although the future of COVID-19 is uncertain, after pausing and learning from its experiences, Holy Cross Health is better prepared for future surges of COVID-19 or other epidemics or disasters.

In lessons from start-ups, Amy Edmondson points out that rapid learning is key to success in highly volatile environments.(10) The PaL process could be easily adapted within other hospitals and healthcare systems to facilitate rapid learning. It could also enable the sharing of meaningful lessons learned among those in the healthcare industry by creating a common framework.

References

1. Chassin MR, Loeb, JM. High-Reliability Health Care: Getting There from Here. Milbank Quarterly. 2013; 91(3): 459–90.

2. Weick, KE, Sutcliffe, KM. Managing the Unexpected – Assuring High Performance in an Age of Complexity. San Francisco, CA: Jossey-Bass; 2001:10–17.

3. Rogers EW, Milam J. Pausing for Learning: Applying the After-Action Review Process at the NASA Goddard Space Flight Center. Big Sky, MT: IEEE Aerospace Conference; 2005: 4383–88.

4. Rogers EW. Introducing the Pause and Learn (PaL) Process. NASA Goddard White Paper. Revised June 8, 2006. www.nasa.gov/centers/goddard/pdf/287922main_PALwhitepaperV3.pdf . Accessed July 28, 2020.

5. Maryland Department of Health. Coronavirus Disease 2019 (COVID-19) Outbreak. 2020. https://coronavirus.maryland.gov . Accessed June 15, 2020.

6. NASA/GSFC. Pause and Learn Implementation Guide. NASA Goddard. July 2011. www.nasa.gov/sites/default/files/files/Pause-and-Learn-Facilitation-Guide.pdf . Accessed July 28, 2020.

7. Agency for Healthcare Research and Quality. Becoming a High Reliability Organization: Operational Advice for Hospital Leaders. Rockville, MD. April 2008; Publication No. 08-0022.

8. Argyris C. Double Loop Learning in Organizations. Harvard Business Review. September-October 1977; 77502:117–24.

9. Argyris C, Schon D. Organizational Learning II: Theory, Method and Practice. Reading, MA: Addison-Wesley; 1996.

10. Edmondson, Amy. What Hospitals Overwhelmed by Covid-19 Can Learn from Startups. Harvard Business Review. May 2, 2020. https://hbr.org/2020/05/what-hospitals-overwhelmed-by-covid-19-can-learn-from-startups . Accessed July 28, 2020.